Method and apparatus for video coding using intra prediction based on template matching

Abstract

A method and an apparatus for video coding using template matching-based intra prediction are disclosed. The video coding method and the video coding apparatus select an intra predictor by defining previously reconstructed pixels spatially adjacent to the current block as a template and performing a template matching-based search in the previously reconstructed region neighboring the current block.

What is claimed is: 
     
         1 . A method performed by a video decoding apparatus for decoding a current block, the method comprising:
 decoding, from a bitstream, an index indicating a target search region that is one of multiple search regions, the search region being a portion of a previously reconstructed region in a current picture;   setting the target search region based on the index;   setting a template from pixels spatially adjacent to the current block;   searching for a similar template with a highest degree of similarity to the template by performing a search based on template matching in the target search region; and   determining a matching block with the current block based on the similar template and selecting the matching block as an intra predictor,   wherein the multiple search regions are formed with both a current virtual pipeline data unit (VPDU), which contains the current block, and three consecutive VPDUs, that precede the current VPDU in decoding order, on a current coding tree unit (CTU) row.   
     
     
         2 . The method of  claim 1 , wherein searching for the similar template comprises:
 searching for the similar template based on a cost function that calculates similarities between the template and candidate templates in the target search region.   
     
     
         3 . The method of  claim 2 , wherein searching for the similar template comprises:
 when the template is larger than a predefined width or a predefined height, calculating a value of the cost function by using pixels obtained by sub-sampling the template and the candidate templates.   
     
     
         4 . The method of  claim 1 , further comprising:
 decoding a residual block from the bitstream; and   reconstructing the current block by adding the intra predictor and the residual block.   
     
     
         5 . A method performed by a video encoding apparatus configured to encode a current block, the method comprising:
 determining an index indicating a target search region that is one of multiple search regions, the search region being a portion of a previously reconstructed region in a current picture;   setting the target search region based on the index;   setting a template from pixels spatially adjacent to the current block;   searching for a similar template with a highest degree of similarity to the template by performing a search based on template matching in the target search region; and   determining a matching block with the current block based on the similar template and selecting the matching block as an intra predictor,   wherein the multiple search regions are formed with both a current virtual pipeline data unit (VPDU), which contains the current block, and three consecutive VPDUs, that precede the current VPDU in encoding order, on a current coding tree unit (CTU) row.   
     
     
         6 . The method of  claim 5 , further comprising:
 generating a residual block by subtracting the intra predictor from the current block; and   encoding the index and the residual block.   
     
     
         7 . A method for providing a video decoding apparatus with video data, the method comprising:
 encoding the video data into a bitstream; and   transmitting the bitstream to the video decoding device,   wherein encoding the video data comprises:
 determining an index indicating a target search region that is one of multiple search regions, the search region being a portion of a previously reconstructed region in a current picture; 
 setting the target search region based on the index; 
 setting a template from pixels spatially adjacent to a current block; 
 searching for a similar template with a highest degree of similarity to the template by performing a search based on template matching in the target search region; and 
 determining a matching block with the current block based on the similar template and selecting the matching block as an intra predictor, 
 wherein the multiple search regions are formed with both a current virtual pipeline data unit (VPDU), which contains the current block, and three consecutive VPDUs, that precede the current VPDU in encoding order, on a current coding tree unit (CTU) row.